nushell/crates/nu-command/tests/commands/base/mod.rs
Andrej Kolčin 0560826414
encode/decode for multiple alphabets (#13428)
Based on the discussion in #13419.


## Description

Reworks the `decode`/`encode` commands by adding/changing the following
bases:

- `base32`
- `base32hex`
- `hex`
- `new-base64`

The `hex` base is compatible with the previous version of `hex` out of
the box (it only adds more flags). `base64` isn't, so the PR adds a new
version and deprecates the old one.

All commands have `string -> binary` signature for decoding and `string
| binary -> string` signature for encoding. A few `base64` encodings,
which are not a part of the
[RFC4648](https://datatracker.ietf.org/doc/html/rfc4648#section-6), have
been dropped.


## Example usage

```Nushell
~/fork/nushell> "string" | encode base32 | decode base32 | decode
string
```

```Nushell
~/fork/nushell> "ORSXG5A=" | decode base32
# `decode` always returns a binary value
Length: 4 (0x4) bytes | printable whitespace ascii_other non_ascii
00000000:   74 65 73 74                                          test
```


## User-Facing Changes

- New commands: `encode/decode base32/base32hex`.
- `encode hex` gets a `--lower` flag.
- `encode/decode base64` deprecated in favor of `encode/decode
new-base64`.
2024-08-23 11:18:51 -05:00

25 lines
574 B
Rust

use data_encoding::HEXUPPER;
use rand::prelude::*;
use rand_chacha::ChaCha8Rng;
mod base32;
mod base32hex;
mod base64;
mod hex;
/// Generate a few random binaries.
pub fn random_bytes() -> Vec<String> {
const NUM: usize = 32;
let mut rng = ChaCha8Rng::seed_from_u64(4);
(0..NUM)
.map(|_| {
let length = rng.gen_range(0..512);
let mut bytes = vec![0u8; length];
rng.fill_bytes(&mut bytes);
let hex_bytes = HEXUPPER.encode(&bytes);
format!("0x[{}]", hex_bytes)
})
.collect()
}